Study On Coke Quality Forecasting And Blast Furnace Optimization Model

The process of coking and sintering and the process of ironmaking in blast furnace are the primary course in the ironmaking and steelmaking process. There are many flows and matters, the complex technological operation in the three processes. In practice the coke quality fluctuates frequently, the scheduling of the section of the sintering to the blast furnace often is disordered. Building up the coke quality forecasting model is favorable to stabilize the coke performance, increase the quantity of the coke. Making up the scheduling optimization model of the section of the sintering to the blast furnace is very useful to steadily product, enhance the productive efficiency. So far, there are many researches about the forecast of coke quality and the process from the sintering to blast furnace and various models are built up. But there are some disadvantages in the published research, such as the low forecasting precision about the coke quality model and there are no mature scheduling models reported or used in this field.In order to solve the problem about the fluctuation of the coke quality and the problem about the scheduling from sintering to blast furnace, this paper builds up three different forecasting models for coke quality through three various methods and a scheduling modeling about the process from the sintering to blast furnace through a method integrated Petri Net and genetic algorithm. The primary content is listed as follow:①The paper proposes the academic background and significance of the subject and explains the significance of the coke quality forecasting model and mass scheduling model. The current situation and development trend of the field at home and abroad have done in the introduction, and the main content and innovative points of the paper have been proposed.②The paper in detail introduces the primary characteristic of the ironmaking process.③In order to solve the problem about the coke quality forecasting, firstly, a model about the relation between the each coal and the mixed coal is built up and three forecasting models are also made using SVM, BP and anfis methods. The results from the three models and actual measured value are compared.④The model integrated Petri Net and GA (genetic algorithm) as suitable tools for simulation and optimization of scheduling to sintering and BF process. The hybrid objective function integrated continuous function about the key operations, the supply function about the plan and the stability function about the key stores is proposed. The best scheduling plan can be found through the genetic algorithm.⑤Based with the practice in the ironmaking plant of Chongqing iron and steel company, the conclusion is drawn that scheduling with optimized values can maintain continuity of key operations, satisfy the plan of hot metal in maximal degree and keep the best stability of key storages.